基于无锚点超宽带的多无人机智能自主相对定位方法及实验验证

In this project, by taking into account sensor noises, uncertainties and no available predefined anchor nodes or markers in unknown environments, an anchor node-free ultra-wideband (UWB) based cooperative relative localization (RL) for multiple unmanned aerial vehicles (UAVs) will be studied. This project aims to provide an applicable solution for multi-UAV collaborative operation in satellite-denied and unstructured environments. Firstly, modelling and estimation of sensor measurement errors and uncertainties are explored. Secondly, without any external infrastructures prepositioned, each UAV estimates the direct and indirect relative positions to its neighbours and cooperatively performs a consensus-based fusion, which fuses the obtained direct and indirect RL estimates, to generate the relative positions in real time despite the fact that some UAVs may not have direct range measurements to their neighbours. Then, based the modelled sensor errors and uncertainties, a compensation method will be proposed and applied into the anchor-node free UWB based cooperative RL method. Finally, extensive simulations and real-world tests will be presented to corroborate the effectiveness of our proposed anchor node-free UWB based cooperative relative localization system. The research topic to be studied is hot and challenging in both self-localization and multi-agent control area which are of great importance in both theory and practice.

本项目面向卫星拒止与非结构化环境下多无人机协同作业的应用需求，针对多无人机系统在复杂未知环境下易受传感器误差、环境不确定性等影响，以及未知环境下无法预设锚点或标志点等问题，基于超宽带测距与通讯技术，旨在系统地研究无锚点多无人机分布式协同相对定位方法，实现多无人机在无锚点非结构化环境下智能自主相对定位的功能。首先，探索多无人机系统传感器误差与不确定性建模估计问题。然后，针对未知环境中无锚点或标志点的问题，提出基于超宽带网络的分布式直接与间接相对位置估计，并通过设计分布式信息融合方法获得无人机间的相对定位信息。再次，基于传感器与不确定性的量化信息设计补偿算法，并将该补偿技术应用到无锚点超宽带多无人机的智能自主相对定位方法中。最后，通过搭建的仿真和多无人机测试平台进行算法和飞行的有效性验证。本项目的研究内容是自主导航和多智能体领域的热门研究课题，具有较强的理论意义和实际意义。

针对旋翼无人机在复杂飞行环境中执行任务的高自主和高安全等需求，瞄准“复杂环境下的自主安全控制”这一重大挑战，研究多无人机卫星拒止环境下协同定位、无人机干扰/故障量化表征、估计与预测、任务重构、安全控制等科学问题，开展旋翼无人机自主导航与安全控制理论及关键技术的研究工作，实现无锚点协同相对定位、对干扰/故障量化表征与建模、基于微循环“小回路”的干扰/故障智能检测、系统能力量化分析与柔性任务重构等关键技术的突破，并构建“干扰/故障-任务+控制-评估”的综合测试验证平台。通过项目的实施，系统性地研究旋翼无人机自主定位与安全控制理论方法，完成核心关键技术的试飞验证，为旋翼无人机的飞行安全保驾护航，提升我国无人机安全控制技术的创新能力。